Anthropic CEO Dario Amodei is reportedly re-engaging in discussions with the Department of Defense, seeking to mend the company's ties with the U.S. military and avoid exclusion from defense contracts due to "supply chain risk" concerns. These negotiations come after talks broke down on Friday, following weeks of intense public disagreement regarding Anthropic's unwillingness to grant the Pentagon unfettered access to its artificial intelligence technology, a void competitors such as OpenAI have been quick to exploit.
Sources familiar with the situation, cited by the Financial Times, indicate that Amodei is currently negotiating with Emil Michael, the Under-Secretary of Defense for Research and Engineering. The aim is to secure a new contract that would permit the U.S. military to continue utilizing Anthropic’s Claude AI models. This follows a heated social media exchange last week where Michael publicly denounced Amodei as a “liar” with a “God-complex,” accusing him of “putting our nation’s safety at risk” amidst a contentious dispute over appropriate military applications of AI.
For the U.S. startup, securing this new agreement may be critical for its continued viability. Secretary of Defense Pete Hegseth had stated his intention on Friday to classify Anthropic as a "supply chain risk." This designation is typically applied to entities linked to foreign governments that could jeopardize U.S. national security. Such a classification would have significant repercussions across the U.S. tech industry, compelling companies to abandon Claude and disassociate from Anthropic if they wish to remain eligible for defense contracts.
Further exacerbating the already strained relationship between Anthropic and the Trump administration is a recently leaked memo, sent by Amodei to his staff on Friday. This internal communication, initially reported by The Information and also reviewed by the Financial Times, reportedly shows Amodei dismissing OpenAI’s agreement with the Pentagon as “safety theater” and characterizing the public statements from both involved parties as “straight up lies.”
Amodei implied that Anthropic’s deteriorating relationship with the federal government stemmed from the company's refusal, unlike OpenAI or its leadership, to "donate to Trump" or offer "dictator-style praise to Trump." OpenAI CEO Sam Altman is recognized among numerous influential figures in Silicon Valley who have cultivated closer ties with the transactional president since his return to office, while OpenAI co-founder and long-time president Greg Brockman is a notable Trump mega-donor.
Within the same memo, Amodei also revealed that the Department of Defense was reportedly nearing acceptance of Anthropic’s proposed terms. He quoted: “Near the end of the negotiation the [department] offered to accept our current terms if we deleted a specific phrase about ‘analysis of bulk acquired data’ which was the single line in the contract that exactly matched this scenario we were most worried about. We found that very suspicious.”
The core of Anthropic’s contentious disagreement with the Pentagon revolves around the Defense Department’s demand for unrestricted access to the company’s technology. Anthropic, however, has steadfastly refused to concede on its two fundamental "red lines" for military application: prohibiting mass surveillance of Americans and the deployment of lethal autonomous weapons—AI systems capable of killing without human oversight. Hegseth has maintained that the AI technology utilized by the department should be available for “any lawful use,” a condition Anthropic has rejected due to fears it could breach these critical boundaries. In contrast, xAI and OpenAI have reportedly accepted these terms.
Until recently, Claude held the distinction of being the sole AI system with the necessary security clearance to process classified information. Available reports suggest that Claude has been actively deployed in military operations, notably in the U.S. raid on Venezuela and for strikes in Iran. These deployments likely influenced President Trump to reverse his previous ban on the federal government’s use of Claude.
The Editorial Staff at AIChief is a team of professional content writers with extensive experience in AI and marketing. Founded in 2025, AIChief has quickly grown into the largest free AI resource hub in the industry.